Google translate instantly translates between any pair of over eighty human languages like French and English. How does it do that? Why does it make the errors that it does? And how can you build something better? Modern translation systems like Google Translate, Bing Translator, and SDL FreeTranslation learn to translate by reading millions of words of already translated text.

This course will show you the rich history behind these systems, and what MT can be used for today.

Instructor

Lane Schwartz - Office hours Tue 5:00-5:30 PM and Thu 8:20-9:00 in Foreign Languages Building, room 4019, and by appointment

Time and place

Tuesdays and Thursdays 2:00-3:20pm, Foreign Languages Building, room G30

Textbook

Most of the readings for this class will be available online. For the remainder of the readings, the following texts will be used:

Statistical Machine Translation (errata) by Philipp Koehn
You can rent or purchase the Kindle eBook version of this text from Amazon or purchase a paper copy from Amazon or from the campus bookstore. The textbook can be read online through the UIUC library, but because the library’s license with the publisher restricts the number of simultaneous users, you should not rely on this as your only mechanism for accessing the book.

The LaTeX Wikibook, The Not So Short Introduction to LaTeX 2e, and Pro Git
In this course you will use LaTeX for typesetting your assignments and Git for version control. Selected readings will come from these three references, all of which are available for free online.

The Linux Command Line: A Complete Introduction by William E. Shotts, Jr.
This course will require some use of the Linux command line. If you do not have prior experience in this area, this supplementary text is strongly recommended. This book is available for free in PDF form directly from the author under a Creative Commons license. If you want a paper, mobi, or epub copy, you can purchase this textbook in both paper and electronic formats directly from the publisher.